KR101829134B1 - Fault prediction system for rotating body using vibration and method thereof - Google Patents

Abstract

The present invention relates to a system and a method for predicting a fault of a rotor, and more specifically, to a system and a method for predicting a rotor using vibration which predict and diagnose failure occurring possibility of a rotor by collecting vibration data on vibration generated by rotation of the rotor, and calculating information entropy after converting the collected vibration data into a frequency domain so as to measure entropy of the rotor.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a vibration prediction system,

The present invention relates to a system and a method for predicting a failure of a rotating body using vibration, and more particularly, to a system and method for estimating a failure of a rotating body by using vibration to collect vibration data on vibration generated by rotation of a rotating body, And more particularly, to a system and method for estimating a failure of a rotating body using vibration to predict and diagnose the possibility of failure of the rotating body by measuring entropy of the rotating body by calculating entropy.

Generally, rotary machines are used to transport and rotate objects in various fields such as factory automation equipment, automobiles, airplanes, and the like, or to rotate propellers and tires.

Such rotating machines include bearings for smoothly rotating objects (propellers, tire wheels, rotating lathes, etc.) to be rotated, automobile propulsion shafts (hereinafter, collectively referred to as "rotating bodies"

If an abnormality occurs in the rotating body, it is highly likely to lead to a major accident. Therefore, maintenance is very important in the field where these rotors are applied. Especially, it will be more important in airplanes.

Conventionally, in order to repair the rotating bodies of such rotating machines, the rotating body must be directly seen by eyes, and therefore, the entire rotating machine must be disassembled.

In order to disassemble the rotating machine as a whole, professional manpower is required. Professional manpower must take time and effort to disassemble the rotating machine to check whether the rotating machine is broken or failed. In addition, there is a problem that the failure of the rotating body can not be recognized by confirming the expert manpower with eyes.

Due to such a problem, Korean Patent Registration No. 10-0228023 (hereinafter referred to as "Prior Patent") discloses a bearing life prediction method for estimating the life of a bearing using vibration.

The prior patents use the spike energy of the vibration signal generated in the bearing to predict the life of the bearing.

However, the prior art has a problem in that it is difficult to analyze the rotors more precisely by predicting the lifetime by using the spike energy of the vibration signal only in the time domain, and the precision is lowered due to this.

Therefore, there is a demand for a method for predicting and diagnosing the life of the rotating body more precisely and accurately.

Patent No. 10-0228023

SUMMARY OF THE INVENTION Accordingly, an object of the present invention is to collect vibration data for vibration generated by rotation of a rotating body, convert the collected vibration data into a frequency domain, calculate information entropy, measure the degree of disorder of the rotating body, And to provide a system and method for estimating a failure of a rotating body using vibration for predicting and diagnosing the occurrence probability of the rotor.

According to an aspect of the present invention, there is provided a system for predicting a failure of a rotating body using vibration, the system including: a cumulative time-domain vibration data cumulatively accumulated in a cycle; a cycle for each frequency included in cumulative time- A storage unit for storing cumulative frequency domain vibration data and cumulative entropy data according to a cycle of the cumulative frequency domain vibration data; A vibration measuring unit connected to the rotating body to measure a vibration generated when the rotating body rotates and to output time-domain vibration data according to the cycle; And a controller for accumulating and storing time-domain vibration data measured through the vibration measuring unit in the storage unit, converting the frequency-domain vibration data into at least one frequency-domain frequency-domain vibration data for the time-domain vibration data measured through the vibration measuring unit, And storing the accumulated frequency-domain frequency-domain vibration data in the storage unit, generating entropy for the frequency-dependent frequency-domain vibration data for each frequency to update the corresponding accumulated entropy data in the storage unit, And determining whether there is a failure in the rotating body according to whether or not the rotating body falls below a reference value.

The control unit may include: a vibration vibration data collection unit for collecting time-domain vibration data according to a cycle through the vibration measurement unit and accumulating the time-domain vibration data in the storage unit; And a frequency domain transforming unit for performing fast Fourier transform on the time domain vibration data according to the cycle collected through the vibration data collecting unit and converting the time domain vibration data into frequency domain vibration data corresponding to at least one frequency included in the time domain vibration data, part; An entropy generation unit for calculating entropy by the frequency-dependent cumulative frequency-domain vibration data and updating accumulated entropy data of the storage unit; And a failure predicting unit for determining whether the entropy value of the accumulated entropy data is continuously decreased or not and determining the possibility of occurrence of a failure according to whether or not the entropy value falls below a reference value.

Wherein the entropy generator selects an upper certain number of entropy reduction among the accumulated entropy data for the at least one frequency and averages the selected number of the accumulated entropy data to convert it into averaging cumulative entropy data, The fault prediction unit may determine whether the entropy values of the averaged accumulation entropy data are decreasing and whether the entropy value falls below a reference value to determine the possibility of failure of the rotor.

The failure predicting unit calculates a failure occurrence rate if the calculated entropy value is less than the previous entropy value and does not fall below the reference value, and notifies the connection rate through the alarm unit.

The entropy generator may calculate an entropy according to the following equation.

[Mathematical Expression]

Where X is the amplitude value of the frequency domain vibration data, n is the number of bins containing time domain vibration data, and p (x _i ) is the probability of each bin. The bin is divided into 255 equal parts from 0 to 1, the time domain vibration data is included in the bin to which the size belongs, and the probability is calculated by dividing the number of vibration data included in bin by the total number.

The warning unit may include at least one of at least one of failure information of the rotating body, measured time domain vibration data, accumulated time domain vibration data, frequency domain vibration data, accumulated frequency domain vibration data, and entropy accumulation data as at least one of text and graphics A display unit for displaying the image; A warning sound generating unit for outputting a possibility of occurrence of a failure of the rotating body as one of a warning sound and a warning sound; And at least one of at least one of failure information of the rotating body, measured time domain vibration data, accumulated time domain vibration data, frequency domain vibration data, accumulated frequency domain vibration data, and entropy accumulation data to a manager terminal of a predetermined administrator And a communication unit.

According to another aspect of the present invention, there is provided a method for predicting a rotating-body failure using vibration, the method including: acquiring time-domain vibration data by accumulating time-domain vibration data measured from a rotating body through a vibration measuring unit, process; A frequency domain transformation step of converting the frequency domain vibration data into at least one frequency domain frequency domain vibration data for the time domain vibration data measured through the vibration measurement part and accumulating the converted frequency domain vibration data in the storage part; An entropy updating step of causing the control unit to generate entropy for accumulated frequency-domain vibration data for each frequency accumulated for each frequency to update corresponding accumulated entropy data of the storage unit; Determining whether a entropy value of the updated accumulated entropy data is decreased and decreasing a entropy value of the updated accumulated entropy data if the entropy value falls below a reference value; And an alarming step of alerting the possibility of the failure of the rotor through the alarm unit when it is determined that the controller has a possibility of failure.

The entropy updating step may include: an updating step of calculating entropy by the frequency-dependent cumulative frequency domain vibration data and updating accumulated entropy data of the storage; A cumulative entropy data selecting step of selecting an upper certain number of entropy reduction widths among the cumulative entropy data for the at least one frequency; And averaging the predetermined number of accumulated entropy data to convert the average accumulated entropy data into averaged cumulative entropy data and storing the average accumulated cumulative entropy data in the storage unit. In the failure occurrence probability determination process, the controller calculates a sum of the entropy values of the averaged cumulative entropy data And determining whether the entropy value falls below a reference value to determine the possibility of the failure of the rotor.

Wherein the failure occurrence probability determination step includes: a decrease determination step of determining whether the entropy value of the updated accumulated entropy data is decreased; A failure occurrence probability monitoring step of determining whether an entropy value of the accumulated entropy data falls below a reference value when the entropy value of the accumulated entropy data decreases; And a failure occurrence possibility determination step of determining that there is a possibility of a failure occurring in the rotor if the entropy value falls below a reference value in the failure occurrence probability monitoring step and that the failure is normal when the entropy value is equal to or greater than a reference value.

Wherein the fault occurrence probability determination step further includes a fault occurrence rate calculation step of calculating the fault occurrence rate of the rotor if the entropy value of the reference value is equal to or greater than a reference value and the fault occurrence rate is output through the alarm part in the alarm procedure .

The entropy is characterized by calculating entropy by the following equation.

[Mathematical Expression]

Where X is the amplitude value of the frequency domain vibration data, n is the number of bins containing time domain vibration data, and p (x _i ) is the probability of each bin. The bin is divided into 255 equal parts from 0 to 1, the time domain vibration data is included in the bin to which the size belongs, and the probability is calculated by dividing the number of vibration data included in bin by the total number.

According to the present invention, it is possible to acquire and analyze more various information by analyzing the vibration data in the time domain by converting it into the frequency domain, and thereby it is possible to predict and diagnose the possibility of occurrence of the failure of the rotating body more precisely .

Further, the present invention calculates the entropy of the vibration data converted into the frequency domain, diagnoses the life of the rotating body and the possibility of occurrence of the fault according to the increase or decrease of the entropy. Therefore, .

1 is a block diagram of a system for predicting a failure of a rotating body using vibration according to the present invention.
2 is a diagram showing signal waveforms by vibration data in a time domain according to test conditions performed according to an embodiment of the present invention.
3 is a diagram for explaining a process of obtaining vibration signal variation data of a specific frequency according to time, which is applied according to an embodiment of the present invention.
4 is a view for explaining the reason why the amplitude magnitude of the frequency domain vibration data by the fast Fourier transform may decrease or increase according to an embodiment of the present invention.
5 is a graph for explaining the concept of entropy change according to the form of vibration data according to an embodiment of the present invention.
6 is a diagram showing signal waveforms and entropy waveforms of frequency domain vibration data for one frequency component according to an embodiment of the present invention.
FIG. 7 is a diagram showing entropy waveforms for seven kinds of rotators in first and second conditions applied in accordance with an embodiment of the present invention. FIG.
8 is a flowchart illustrating a method for predicting a failure of a rotating body using vibration according to the present invention.
9 is a flowchart illustrating a method for determining the probability of occurrence of a fault in the method for predicting the failure of a rotating body using vibration according to the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will now be described in detail with reference to the accompanying drawings.

FIG. 1 is a diagram illustrating a configuration of a system for predicting a failure of a rotating body using vibration according to an embodiment of the present invention. FIG. 2 is a graph showing waveforms of vibration data in a time domain according to a test condition performed according to an embodiment of the present invention. FIG. 3 is a diagram for explaining a process of obtaining vibration signal variation data of a specific frequency according to time, which is applied in accordance with an embodiment of the present invention. FIG. FIG. 5 is a graph for explaining the concept of entropy increase / decrease according to the form of vibration data according to an embodiment of the present invention 6 is a graph showing a signal waveform and an entropy waveform of frequency domain vibration data for one frequency component according to an embodiment of the present invention And FIG. 7 is a view showing entropy waveforms for seven kinds of rotating bodies under the first condition and the second condition applied according to an embodiment of the present invention. Hereinafter, a description will be given with reference to Figs. 1 to 7. Fig.

The system for estimating the failure of a rotating machine using vibration according to the present invention includes a storage unit 10, an alarm unit 20, a vibration measurement unit 30, and a control unit 40.

The storage unit 10 accumulates time-domain vibration data, which is vibration data in the time domain, according to the cycle (hereinafter, time-domain vibration data stored cumulatively in accordance with the cycle is referred to as "cumulative time-domain vibration data" (Hereinafter, referred to as "cumulative frequency domain vibration data" stored cumulatively in accordance with a cycle) is stored in the frequency domain vibration storage 11, And an entropy storage unit 13 for accumulating and storing entropy according to a cycle for the data storage unit 12 and the individual frequency vibration data (hereinafter, entropy stored cumulatively in accordance with the cycle is referred to as "cumulative entropy data").

The alarm unit 20 is a means for alerting the possibility of occurrence of a failure of the rotating body under the control of the control unit 40 to determine the possibility of a failure, A warning sound generating unit 22 for outputting a possibility of occurrence of a failure of the rotating body by voice or warning sound, a wired / wireless communication unit (not shown) for transmitting a message for notifying the manager of the remote place to the possibility of occurrence of a failure of the rotating body 23). The message transmitted through the wired / wireless communication unit 23 may be a mobile communication message such as an e-mail, a push message, or a short message service (SMS) message. The administrator terminal may be a computer, a notebook, a smart phone, a mobile phone, or the like.

The vibration measuring unit 30 includes a vibration sensor (not shown) attached to the rotating body to output a vibration signal corresponding to the vibration of the rotating body to determine the possibility of failure, Converts the vibration signal in the time domain into the time domain vibration data, and outputs the time domain vibration data. The rotating body may be a bearing, a motor, or the like, and may be any rotating object.

The time domain vibration data output from the vibration measurement unit 30 may be as shown in FIG. Fig. 2 (A) shows the result of measurement on seven types of bearings in the first condition of Table 1, and Fig. 2 (B) shows the measurement of seven kinds of bearings in the second condition of Table 1 below. In FIG. 2, the time domain vibration data, the frequency domain vibration data, and the cumulative entropy data are classified into 1 set, 2 sets, 3 sets, 4 sets, 5 sets, 6 sets and 7 sets according to the types of bearings.

For example, the first bearing (first set) of FIG. 1 shows that the vibration becomes larger at 2800 cycles in the first condition and larger at 900 cycles in the second condition.

The control unit 40 controls the overall operation according to the present invention and in particular accumulates the time domain vibration data output from the vibration measurement unit 30 in the time domain vibration data storage unit 11 of the storage unit 10 , And converts the time-domain vibration data into frequency-domain vibration data, which is frequency-domain data in accordance with the cycle, and accumulates the frequency-domain vibration data in the frequency-domain vibration data storage unit 12 of the storage unit 10, And accumulates the entropy in the entropy storage unit 13 and stores the entropy.

The control unit 40 analyzes accumulated entropy data accumulated in the entropy storage unit 13 of the storage unit 10 to determine whether the entropy value of the accumulated entropy data is decreasing. If the entropy value is decreasing, If the vibration entropy value is less than the reference value, it is determined that there is a possibility of failure in the rotating body, and the alarm unit 20 is alarmed that there is a possibility of a failure in the rotating body.

The control unit 40 includes a vibration data collecting unit 41, a frequency domain transforming unit 42, an entropy generating unit 43, and a failure predicting unit 44 do.

The vibration data collecting unit 41 collects time-domain vibration data for a predetermined period of time at regular intervals with respect to the vibration of the rotating body through the vibration measuring unit 30, and accumulates the collected time-domain vibration data in the time- For example, the time may be 1/10 of a second as shown in FIG. 3 (a), and the period may be 10 seconds. In addition, the sampling frequency for collecting the time domain vibration data may be 25.6 KHz.

The frequency domain transforming unit 42 receives the time domain vibration data according to the cycle through the vibration data collecting unit 41 and performs fast Fourier transform (FFT) on the time domain vibration data, And outputs the frequency-domain vibration data.

As shown in (B) and (D) of FIG. 3, time-domain vibration data having at least one frequency is combined with the time-domain vibration data as shown in (A) of FIG.

4, time-domain vibration data, which is raw data such as (a) in FIG. 4, output from the vibration measuring unit 30, is obtained by combining time-domain vibration data of frequency A and frequency B But can not be confirmed in the time domain.

Accordingly, when the time-domain vibration data is converted into the frequency domain through the frequency domain transformer 42, the time-domain vibration data includes data of the frequency components of the frequency A and the frequency B as shown in (C) of FIG. 4 And converts the frequency domain vibration data into frequency domain vibration data for each frequency component included in the time domain vibration data as shown in (b) of FIG.

The entropy generator 43 receives the FFT-transformed frequency-domain vibration data from the frequency-domain transformer 42 according to the cycle for each frequency and applies the frequency-domain frequency-domain vibration data to the following equation (1) (Entropy) H (X), accumulates the calculated entropy in the entropy storage unit 10, and updates the accumulated entropy data in the storage unit 13. The entropy (H (X)) of the present invention is information entropy, indicating the congestion of the system, and more clearly indicating the state of the rotating body than the amplitude of the vibration data.

Where X is the amplitude value of the frequency domain vibration data, n is the number of bins containing time domain vibration data, and p (x _i ) is the probability of each bin. 5, the bin is divided into 255 equal parts from 0 to 1, the time domain vibration data is included in the bin to which the size belongs, and the probability is a value obtained by dividing the number of vibration data included in bin by the total number .

The entropy H (X) is increased when the time domain vibration data is formed in the diffusion type as shown in (A) of FIG. 5, and the time domain vibration data is converged as shown in (B) Decreases when formed.

The entropy generator 43 may obtain the entropy waveform as shown in (B) of FIG. 6 when the above-described Equation 1 is applied to the frequency domain vibration data having one frequency component as shown in (A) of FIG. 6 .

Also, the entropy generator 43 selects a predetermined fixed number of the accumulated entropy data in descending order of the decreasing width of the accumulated entropy data for at least one frequency component, and calculates the average accumulated entropy data obtained by averaging the selected accumulated entropy data And store it in the entropy storage unit 13 of the storage unit 10. For example, FIG. 7 (a) shows the top 25 frequency components in time, indicating that the top 25 frequency components converge to a similar size after a certain time, and FIG. 7 (B) Shows the average cumulative entropy data obtained by averaging the cumulative entropy data for the 25 frequency components, and graphically shows the average cumulative entropy data for seven types (seven sets) of bearings.

The failure predicting unit 44 loads the accumulated entropy data stored in the entropy storage unit 13 every predetermined period, analyzes the loaded accumulated entropy data, determines the possibility of occurrence of a failure of the corresponding rotor, And alerts the manager of the possibility judgment result through the alarm unit 20. [

When the average accumulated entropy data is generated, the failure predicting unit 44 is preferably configured to determine the possibility of occurrence of the failure of the corresponding rotating body by the average accumulated entropy data.

6, when the minimum entropy is lower than a preset reference value, it is determined that there is a possibility of failure in the rotating body, Lt; RTI ID = 0.0 > fault. ≪ / RTI >

8 is a flowchart illustrating a method for predicting a failure of a rotating body using vibration according to the present invention.

Referring to FIG. 8, when the rotor monitoring event occurs, the control unit 40 collects and accumulates vibration data for the subject rotation body (S111). The vibration data includes time domain vibration data, frequency domain vibration data, and entropy. The rotating body monitoring event may be generated when the system including the rotating body is driven.

When the vibration data for the rotating body is collected and accumulated, the controller 40 analyzes the accumulated entropy data for the rotating body (S113), and determines whether there is a possibility that the rotating body is normal or a failure occurs according to the analyzed result (S115).

If it is normal, the control unit 40 alerts the alarming unit 20 via the alarm unit 20 (S117), and warns that failure may occur if a failure occurs (S119). The control unit 40 may calculate the failure occurrence rate and display the failure occurrence rate through the display unit 21. [

FIG. 9 is a flowchart showing a method for determining the probability of occurrence of a failure in the method for predicting the failure of a rotating body using vibration according to the present invention. FIG. 9 is a flow chart showing a method for determining a failure occurrence probability S111 and a rotating body vibration data analysis routine S113 Fig. A method for determining the possibility of occurrence of rotor failure will be described in detail with reference to FIG.

First, the control unit 40 collects time-domain vibration data through the vibration measuring unit 30 at a predetermined cycle (cycle-by-cycle) through the vibration data collecting unit 41 (S211).

When the time domain vibration data is collected, the controller 40 converts the frequency domain vibration data into the frequency domain vibration data through the frequency domain transformer 42 (S213).

When the frequency domain vibration data is generated and stored, the controller 40 calculates the entropy for each frequency domain vibration data through the entropy calculator 43 and updates the accumulated entropy data for each frequency of the entropy storage unit 13 (S215).

After updating the accumulated entropy data, the controller 40 determines whether there is decremented accumulated entropy data among the accumulated entropy data for each frequency component included in the time-domain vibration data through the failure predicting unit 44 (S217) .

If it is determined that the entropy value of the accumulated entropy data is increased through the failure predicting unit 44, the control unit 40 determines that the target rotation is normal (S218).

However, if it is determined that there is accumulated entropy data whose value decreases, the control unit 40 determines that the largest width reduction of entropy sets for a plurality of frequencies as shown in FIG. 7 (a) Entropy sets of the visible upper schedule number (e.g., 25) are selected (S217). At this time, the controller 40 generates average accumulated entropy data obtained by averaging the accumulated entropy data for the one or more frequency components.

When the average accumulated entropy data is generated, the controller 40 determines whether the entropy value of the average accumulated entropy data falls below a preset final reference value through the failure predicting unit 44 at step S221.

If the entropy value does not fall below the reference value, the control unit 40 calculates the failure occurrence rate through the failure prediction unit 44 (S223). The fault occurrence rate may be calculated by the following equation (2).

Where H _ds = the entropy value of the entropy set reduction start, H _C is the entropy value of the current cycle, and H _R is the reference value.

7 (b), the failure occurrence rate is {(3.8-2.6) / (3.8-2.5)} * 100 = 1.2 / 1 in 1800 cycles of set 1 (Red) 1.3 * 100 = 92%.

However, if the entropy value of the decreasing accumulated entropy data is less than the reference value, the control unit 40 determines that the target rotating body has a high possibility of failure (S225).

While the present invention has been described with reference to exemplary embodiments, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, but, on the contrary, is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims. It will be easily understood. It is to be understood that the invention is not limited to the disclosed embodiments, but, on the contrary, it is intended to cover various modifications within the scope of the appended claims.

10:
11: Time domain vibration data storage unit
12: Frequency domain vibration data storage unit
13: Entropy storage unit
20: Alarm section
21: display unit 22: warning sound generating unit
23: wired / wireless communication unit 30: vibration measurement unit
40: control unit 41: vibration data collecting unit
42: frequency domain conversion unit 43: entropy generation unit
44:

Claims (12)

Cumulative frequency domain vibration data according to a cycle for each frequency included in the cumulative time domain data according to the cycle and cumulative entropy data according to a cycle for the cumulative frequency domain vibration data, A storage unit for storing the data;
A vibration measuring unit connected to the rotating body to measure a vibration generated when the rotating body rotates and to output time-domain vibration data according to the cycle; And
Wherein the time-domain vibration data measured through the vibration measurement unit is accumulated in the storage unit and is converted into at least one frequency-domain frequency-domain vibration data for the time-domain vibration data measured through the vibration measurement unit, Frequency-domain frequency-domain vibration data to the storage unit, generates entropy for the frequency-dependent frequency-domain vibration data for each frequency to update the corresponding accumulated entropy data in the storage unit, and determines whether the updated accumulated entropy data is decreased And determining whether or not a failure occurs in the rotating body according to whether or not the rotating body falls below a reference value.
The method according to claim 1,
Wherein,
A vibration data collecting unit collecting time-domain vibration data according to the cycle through the vibration measuring unit and accumulating the time-domain vibration data in the storage unit;
And a frequency domain transforming unit for performing fast Fourier transform on the time domain vibration data according to the cycle collected through the vibration data collecting unit and converting the time domain vibration data into frequency domain vibration data corresponding to at least one frequency included in the time domain vibration data, part;
An entropy generation unit for calculating entropy by the frequency-dependent cumulative frequency-domain vibration data and updating accumulated entropy data of the storage unit; And
Determining whether the entropy value of the cumulative entropy data continuously decreases or not and determining a possibility of failure according to whether or not the entropy value falls below a reference value.
3. The method of claim 2,
Wherein the entropy-
Selecting a higher certain number of entropy reduction widths among the accumulated entropy data for the at least one frequency, averaging the selected number of the accumulated entropy data, converting the averaged accumulated entropy data into averaged accumulated entropy data,
Wherein the failure prediction unit comprises:
And determining whether or not the entropy values of the averaged cumulative entropy data decrease and the entropy value of the averaged cumulative entropy data fall below a reference value to determine the possibility of failure of the rotator. system.
3. The method of claim 2,
Wherein the failure prediction unit comprises:
If the entropy value of the calculated accumulated entropy data is less than the entropy value of the previous accumulated entropy data and does not fall below the reference value, the failure occurrence rate is calculated and the failure occurrence rate is notified through the alarm section. system.
3. The method of claim 2,
Wherein the entropy-
And the entropy is calculated by the following equation (3).
&Quot; (3) "

Where X is the amplitude value of the frequency domain vibration data, n is the number of bins containing time domain vibration data, and p (x _i ) is the probability of each bin. The bin is divided into 255 equal parts from 0 to 1, the time domain vibration data is included in the bin to which the size belongs, and the probability is calculated by dividing the number of vibration data included in bin by the total number.
The method according to claim 1,
Further comprising an alarm unit for alerting a possibility of occurrence of a failure of the rotor,
Wherein the control unit generates an alarm through the alarm unit when the occurrence of the fault is determined.
The method according to claim 6,
The alarm unit,
At least one of at least one of fault information of the rotating body, measured time domain vibration data, accumulated time domain vibration data, frequency domain vibration data, accumulated frequency domain vibration data, and entropy accumulation data by at least one of text and graphics ;
A warning sound generating unit for outputting a possibility of occurrence of a failure of the rotating body as one of a warning sound and a warning sound; And
A wired / wireless communication unit for transmitting at least one of fault information of the rotating body, measured time domain vibration data, accumulated time domain vibration data, frequency domain vibration data, accumulated frequency domain vibration data and entropy accumulation data to a manager terminal of a predetermined administrator Wherein the at least one of the at least one rotor and the at least one of the at least one rotor and the at least one rotor is a rotor.
A time domain vibration data acquiring step of accumulating and storing time domain vibration data measured from a rotating body through a vibration measuring unit in a storage unit;
A frequency domain transformation step of converting the frequency domain vibration data into at least one frequency domain frequency domain vibration data for the time domain vibration data measured through the vibration measurement part and accumulating the converted frequency domain vibration data in the storage part;
An entropy updating step of causing the control unit to generate entropy for accumulated frequency-domain vibration data for each frequency accumulated for each frequency to update corresponding accumulated entropy data of the storage unit;
The control unit determines whether or not the entropy value of the updated accumulated entropy data decreases and determines whether or not the entropy value of the accumulated entropy data falls below a reference value, process; And
And an alarming step of alerting the possibility of occurrence of a failure of the rotating body through an alarm part when it is determined that the controller has a possibility of occurrence of a failure.
9. The method of claim 8,
The entropy updating process includes:
An updating step of calculating an entropy (value) by the frequency-dependent cumulative frequency-domain vibration data and updating accumulated entropy data of the storage;
A cumulative entropy data selecting step of selecting an upper certain number of entropy reduction widths among the cumulative entropy data for the at least one frequency; And
And averaging the predetermined number of accumulated entropy data to convert the accumulated accumulated entropy data into averaged cumulative entropy data and storing the same in the storage unit,
The controller determines whether or not the entropy values of the averaged cumulative entropy data are decreasing and whether the entropy value of the averaged cumulative entropy data falls below a reference value to determine the possibility of the occurrence of a fault in the rotational body Wherein the vibration is generated by the vibration of the rotor.
9. The method of claim 8,
The fault occurrence probability determination process includes:
Determining whether the entropy value of the updated accumulated entropy data decreases;
A failure occurrence probability monitoring step of determining whether an entropy value of the accumulated entropy data falls below a reference value when the entropy value of the accumulated entropy data decreases; And
And a failure occurrence possibility determination step of determining that there is a possibility of a failure in the rotor if the entropy value of the accumulated entropy data falls below a reference value in the failure occurrence possibility monitoring step, A Method for Predicting Rotor Faults Using Vibration.
11. The method of claim 10,
The fault occurrence probability determination process includes:
Further comprising a failure occurrence rate calculation step of calculating a failure occurrence rate of the rotor if the entropy value of the accumulated entropy data is equal to or greater than a reference value,
And outputting the fault occurrence rate through the alarm unit in the alarming process.
10. The method according to claim 8 or 9,
The entropy,
Wherein the entropy is calculated by the following equation (4).
&Quot; (4) "

Where X is the amplitude value of the frequency domain vibration data, n is the number of bins containing time domain vibration data, and p (x _i ) is the probability of each bin. The bin is divided into 255 equal parts from 0 to 1, the time domain vibration data is included in the bin to which the size belongs, and the probability is calculated by dividing the number of vibration data included in bin by the total number.

Images